-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  C:\Users\knm4642\Dropbox\Nha Meo Heo Beo\_JMP\WORKING FOLDER\Table A7_Retirement and death events.log
  log type:  text
 opened on:  18 Feb 2026, 23:51:57

. 
. //PREPARE DATA
. use "CEO transition event sample.dta", clear

. * construct additional variables
. gen postxdeltatrust = postchange * deltatrust if trust_sd != .
(111 missing values generated)

. gen postxtrustbf = postchange * trustbf  if trust_sd != .
(111 missing values generated)

. * set globals
. global firm = "firmage firmage2"

. global ceo = "gender age age2 yrinco i.education"

. global eventsample = "!bothnonUS"

. global cluster = "mainethcode"

. 
. 
. //PREPARE TABLE
. foreach panel in "A" "B" {
  2.         * PANEL A: Including all years in each event
.         * PANEL B: Excluding transition years
.         if "`panel'" == "B" {
  3.                 preserve
  4.                 bysort eventid: egen tempendyr = max(termendyr) if postchange == 0
  5.                 bysort eventid: egen starttermendyr = mean(tempendyr)
  6.                 drop if year == starttermendyr & postchange == 1
  7.         }
  8.         
.         eststo clear
  9.         eststo col1: /// retired at 65
>                 reghdfe ash_f1allpat trust_sd ${firm} ${ceo} if ${eventsample} ///
>                 & retire1yr, a(eventid year) cluster(${cluster}) keepsin
 10.                         estadd local sample     "Retd 65"
 11.                         estadd local spec       "OLS"
 12.                         estadd local FE         "X"
 13.                         estadd local controls   "X"
 14.                         distinct eventid if e(sample)
 15.                         eststo col1, add(noevents r(ndistinct))
 16.         eststo col2: /// retired between 64-66
>                 reghdfe ash_f1allpat trust_sd ${firm} ${ceo} if ${eventsample} ///
>                 & retire3yr, a(eventid year) cluster(${cluster}) keepsin
 17.                         estadd local sample     "Retd 64-66"
 18.                         estadd local spec       "OLS"
 19.                         estadd local FE         "X"
 20.                         estadd local controls   "X"
 21.                         distinct eventid if e(sample)
 22.                         eststo col2, add(noevents r(ndistinct))
 23.         eststo col3: /// retired between 63-67
>                 reghdfe ash_f1allpat trust_sd ${firm} ${ceo} if ${eventsample} ///
>                 & retire5yr, a(eventid year) cluster(${cluster}) keepsin
 24.                         estadd local sample     "Retd 63-67"
 25.                         estadd local spec       "OLS"
 26.                         estadd local FE         "X"
 27.                         estadd local controls   "X"
 28.                         distinct eventid if e(sample)
 29.                         eststo col3, add(noevents r(ndistinct))
 30.         eststo col4: /// died within transition year
>                 reghdfe ash_f1allpat trust_sd ${firm} ${ceo} if ${eventsample} ///
>                 & deathwi0yr, a(eventid year) cluster(${cluster}) keepsin
 31.                         estadd local sample     "Died trans yr"
 32.                         estadd local spec       "OLS"
 33.                         estadd local FE         "X"
 34.                         estadd local controls   "X"
 35.                         distinct eventid if e(sample)
 36.                         eststo col4, add(noevents r(ndistinct))
 37.         eststo col5: /// died within transition year + 1
>                 reghdfe ash_f1allpat trust_sd ${firm} ${ceo} if ${eventsample} ///
>                 & deathwi1yr, a(eventid year) cluster(${cluster}) keepsin
 38.                         estadd local sample     "Died trans yr+1"
 39.                         estadd local spec       "OLS"
 40.                         estadd local FE         "X"
 41.                         estadd local controls   "X"
 42.                         distinct eventid if e(sample)
 43.                         eststo col5, add(noevents r(ndistinct))
 44.         eststo col6: /// retired or died OLS
>                 reghdfe ash_f1allpat trust_sd ${firm} ${ceo} if ${eventsample} ///
>                 & (retire5yr | deathwi1yr), a(eventid year) cluster(${cluster}) keepsin
 45.                         estadd local sample     "Retd/died"
 46.                         estadd local spec       "OLS"
 47.                         estadd local FE         "X"
 48.                         estadd local controls   "X"
 49.                         distinct eventid if e(sample)
 50.                         eststo col6, add(noevents r(ndistinct))
 51.         eststo col7: /// retired or died RF
>                 reghdfe ash_f1allpat postxtrustbf postchange ${firm} ${ceo} if ${eventsample} ///
>                 & (retire5yr | deathwi1yr), a(eventid year) cluster(mainethcoder) keepsin
 52.                         estadd local sample     "Retd/died"
 53.                         estadd local spec       "RF"
 54.                         estadd local FE         "X"
 55.                         estadd local controls   "X"
 56.                         distinct eventid if e(sample)
 57.                         eststo col7, add(noevents r(ndistinct))
 58.         eststo col8: /// retired or died IV
>                 ivreghdfe ash_f1allpat (postxdeltatrust = postxtrustbf) postchange ${firm} ${ceo} if ${eventsample} ///
>                 & (retire5yr | deathwi1yr), a(eventid year) cluster(mainethcoder) keepsin first
 59.                         estadd local sample     "Retd/died"
 60.                         estadd local spec       "IV"
 61.                         estadd local FE         "X"
 62.                         estadd local controls   "X"
 63.                         distinct eventid if e(sample)
 64.                         eststo col8, add(noevents r(ndistinct))
 65.         cap restore             
 66.         
.         global coeflist = "trust_sd postxtrustbf postxdeltatrust"
 67.         esttab /*using "Table A7_Retirement and death events_Panel `panel'.tex"*/, ///
>                 cells(b(star fmt(3)) se(par fmt(3))) starlevels(* 0.1 ** 0.05 *** 0.01) ///
>                 nomtitle collabels(none) label varwidth(35) modelwidth(13) ///
>                 mgroups("arsinh(Future patent applications)" "\textbf{D(.)}", pattern(1 0 0 0 0 0 0 0) ///
>                         prefix(\multicolumn{@span}{c}{) suffix(}) span) ///
>                 keep(${coeflist}) order(${coeflist}) ///
>                 coeflab(trust_sd "CEO's trust" ///
>                                 trustbf "Predecessor CEO's trust" ///
>                                 postxtrustbf "Post $\times$ Preceding CEO's trust" ///
>                                 postxdeltatrust "Post $\times$ Change in trust") ///
>                 stats(sample spec FE controls N noevents, fmt(%9.0fc %9.0fc) ///
>                         lab("Sample" "Specification" "Event \& Year FEs" "Baseline controls" "Observations" "Events"))
 68. }
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)
note: firmage is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)

HDFE Linear regression                            Number of obs   =        914
Absorbing 2 HDFE groups                           F(   9,     23) =       1.85
Statistics robust to heteroskedasticity           Prob > F        =     0.1126
                                                  R-squared       =     0.9314
                                                  Adj R-squared   =     0.9217
                                                  Within R-sq.    =     0.0227
Number of clusters (mainethcode) =         24     Root MSE        =     0.5565

                           (Std. err. adjusted for 24 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .2409278   .0926471     2.60   0.016     .0492726     .432583
     firmage |          0  (omitted)
    firmage2 |   -.000385   .0002572    -1.50   0.148    -.0009172    .0001471
      gender |  -.2813492   .2964387    -0.95   0.352    -.8945794     .331881
         age |   .0356231   .0647358     0.55   0.587    -.0982932    .1695394
        age2 |  -.0003425   .0006027    -0.57   0.575    -.0015893    .0009043
      yrinco |   .0047443   .0033749     1.41   0.173    -.0022372    .0117258
             |
   education |
          2  |   .2367723   .1073114     2.21   0.038     .0147817     .458763
          3  |   .3667267   .1393736     2.63   0.015     .0784104    .6550429
          4  |    .345215   .1698496     2.03   0.054    -.0061456    .6965757
             |
       _cons |   -.170887   1.988081    -0.09   0.932    -4.283545    3.941771
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |        93           0          93     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd 65"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |        914         93
(e(noevents) = 93 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =      2,287
Absorbing 2 HDFE groups                           F(  10,     26) =       5.31
Statistics robust to heteroskedasticity           Prob > F        =     0.0003
                                                  R-squared       =     0.9368
                                                  Adj R-squared   =     0.9290
                                                  Within R-sq.    =     0.0073
Number of clusters (mainethcode) =         27     Root MSE        =     0.5321

                           (Std. err. adjusted for 27 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .0686392   .0414009     1.66   0.109    -.0164615      .15374
     firmage |   -.017913   .0058699    -3.05   0.005    -.0299788   -.0058472
    firmage2 |  -.0002098   .0001543    -1.36   0.186     -.000527    .0001075
      gender |  -.1667577   .1740122    -0.96   0.347     -.524445    .1909296
         age |   .0088365   .0355066     0.25   0.805    -.0641483    .0818213
        age2 |  -.0000663   .0003209    -0.21   0.838     -.000726    .0005934
      yrinco |   .0017855   .0019612     0.91   0.371    -.0022458    .0058168
             |
   education |
          2  |   .1778325   .1166827     1.52   0.140    -.0620121    .4176772
          3  |   .1210732   .1212051     1.00   0.327    -.1280674    .3702138
          4  |   .1562387     .14021     1.11   0.275     -.131967    .4444443
             |
       _cons |   1.607264   1.147638     1.40   0.173    -.7517395    3.966267
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       232           0         232     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd 64-66"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       2287        232
(e(noevents) = 232 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =      3,442
Absorbing 2 HDFE groups                           F(  10,     27) =       5.13
Statistics robust to heteroskedasticity           Prob > F        =     0.0003
                                                  R-squared       =     0.9322
                                                  Adj R-squared   =     0.9240
                                                  Within R-sq.    =     0.0050
Number of clusters (mainethcode) =         28     Root MSE        =     0.5354

                           (Std. err. adjusted for 28 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .0707903    .044798     1.58   0.126    -.0211276    .1627082
     firmage |  -.0173154   .0063825    -2.71   0.011    -.0304112   -.0042196
    firmage2 |  -.0000581   .0000772    -0.75   0.458    -.0002164    .0001003
      gender |  -.1200232   .1196981    -1.00   0.325    -.3656235    .1255771
         age |   .0087503   .0253341     0.35   0.732    -.0432311    .0607316
        age2 |  -.0000871   .0002419    -0.36   0.722    -.0005835    .0004093
      yrinco |   .0017532   .0014086     1.24   0.224     -.001137    .0046433
             |
   education |
          2  |   .1733848   .1032153     1.68   0.105    -.0383956    .3851651
          3  |   .1228372   .0980961     1.25   0.221    -.0784394    .3241138
          4  |   .2282021   .1285284     1.78   0.087    -.0355163    .4919206
             |
       _cons |   1.328344    .657896     2.02   0.054    -.0215475    2.678235
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       348           0         348     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd 63-67"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3442        348
(e(noevents) = 348 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 8 iterations)
note: firmage is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)

HDFE Linear regression                            Number of obs   =        253
Absorbing 2 HDFE groups                           F(   9,     11) =    5052.09
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.9335
                                                  Adj R-squared   =     0.9153
                                                  Within R-sq.    =     0.0761
Number of clusters (mainethcode) =         12     Root MSE        =     0.5658

                           (Std. err. adjusted for 12 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .5254958   .2555904     2.06   0.064    -.0370548    1.088046
     firmage |          0  (omitted)
    firmage2 |   .0006803   .0003782     1.80   0.100    -.0001521    .0015127
      gender |   -.533554   .5271395    -1.01   0.333     -1.69378    .6266722
         age |  -.0404415   .0845395    -0.48   0.642    -.2265117    .1456287
        age2 |   .0003359   .0006691     0.50   0.626    -.0011369    .0018087
      yrinco |  -.0308322    .013734    -2.24   0.046    -.0610606   -.0006038
             |
   education |
          2  |   .4109029   .1266881     3.24   0.008     .1320642    .6897416
          3  |   .2485568   .1098848     2.26   0.045     .0067019    .4904116
          4  |   .6794392   .1117745     6.08   0.000     .4334252    .9254531
             |
       _cons |  -.0722041   4.256365    -0.02   0.987    -9.440401    9.295993
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |        34           0          34     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Died trans yr"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |        253         34
(e(noevents) = 34 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 8 iterations)
note: firmage is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)

HDFE Linear regression                            Number of obs   =        353
Absorbing 2 HDFE groups                           F(   9,     12) =      23.39
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.9310
                                                  Adj R-squared   =     0.9151
                                                  Within R-sq.    =     0.0517
Number of clusters (mainethcode) =         13     Root MSE        =     0.5224

                           (Std. err. adjusted for 13 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .5118713   .2763647     1.85   0.089    -.0902756    1.114018
     firmage |          0  (omitted)
    firmage2 |   .0005506   .0003726     1.48   0.165    -.0002611    .0013623
      gender |  -.3866695    .185218    -2.09   0.059    -.7902249    .0168858
         age |  -.0733485   .0717437    -1.02   0.327    -.2296647    .0829676
        age2 |   .0005994   .0005771     1.04   0.319     -.000658    .0018568
      yrinco |  -.0241966   .0114351    -2.12   0.056    -.0491115    .0007183
             |
   education |
          2  |   .3925296    .103765     3.78   0.003     .1664452    .6186141
          3  |    .149743   .0966399     1.55   0.147    -.0608172    .3603032
          4  |   .4527087   .1113998     4.06   0.002     .2099894     .695428
             |
       _cons |    .701267   3.763684     0.19   0.855    -7.499096     8.90163
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |        46           0          46     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Died trans yr+1"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |        353         46
(e(noevents) = 46 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =      3,758
Absorbing 2 HDFE groups                           F(  10,     27) =       7.26
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.9317
                                                  Adj R-squared   =     0.9233
                                                  Within R-sq.    =     0.0056
Number of clusters (mainethcode) =         28     Root MSE        =     0.5361

                           (Std. err. adjusted for 28 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .0848849   .0425261     2.00   0.056    -.0023714    .1721412
     firmage |  -.0194418   .0061031    -3.19   0.004    -.0319643   -.0069192
    firmage2 |  -.0000179   .0000679    -0.26   0.794    -.0001571    .0001214
      gender |  -.1115255   .1078749    -1.03   0.310    -.3328665    .1098154
         age |   .0043571   .0115194     0.38   0.708    -.0192787     .027993
        age2 |  -.0000527   .0001092    -0.48   0.633    -.0002767    .0001713
      yrinco |   .0013758    .001349     1.02   0.317    -.0013921    .0041436
             |
   education |
          2  |   .2127493   .0827792     2.57   0.016     .0429004    .3825983
          3  |   .1683189   .0822764     2.05   0.051    -.0004984    .3371362
          4  |   .2897445   .1014536     2.86   0.008      .081579    .4979101
             |
       _cons |   1.332825   .3740119     3.56   0.001     .5654155    2.100233
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       388           0         388     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd/died"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3758        388
(e(noevents) = 388 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =      3,758
Absorbing 2 HDFE groups                           F(  11,     50) =       5.51
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.9316
                                                  Adj R-squared   =     0.9233
                                                  Within R-sq.    =     0.0053
Number of clusters (mainethcoder) =         51    Root MSE        =     0.5363

                          (Std. err. adjusted for 51 clusters in mainethcoder)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
postxtrustbf |  -.0775358   .0431389    -1.80   0.078    -.1641829    .0091113
  postchange |   .3279969    .217726     1.51   0.138    -.1093186    .7653124
     firmage |  -.0216703   .0052653    -4.12   0.000    -.0322461   -.0110945
    firmage2 |  -2.24e-06   .0001118    -0.02   0.984    -.0002268    .0002223
      gender |  -.1208518   .0898128    -1.35   0.185     -.301246    .0595424
         age |   .0038154    .014167     0.27   0.789    -.0246399    .0322707
        age2 |  -.0000734   .0001243    -0.59   0.557    -.0003232    .0001763
      yrinco |     .00115    .001129     1.02   0.313    -.0011177    .0034177
             |
   education |
          2  |    .223763   .1026997     2.18   0.034     .0174845    .4300415
          3  |   .1830532   .1031304     1.77   0.082    -.0240902    .3901967
          4  |   .2903863   .0910937     3.19   0.002     .1074193    .4733533
             |
       _cons |   1.947089   .4997724     3.90   0.000     .9432669    2.950912
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       388           0         388     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd/died"

added macro:
               e(spec) : "RF"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3758        388
(e(noevents) = 388 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

First-stage regressions
-----------------------


First-stage regression of postxdeltatrust:

Statistics robust to heteroskedasticity and clustering on mainethcoder
Number of obs =                   3758
Number of clusters (mainethcoder) =     51
------------------------------------------------------------------------------
             |               Robust
postxdelta~t | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
postxtrustbf |  -.9400056   .0303793   -30.94   0.000    -.9995694   -.8804418
  postchange |   4.761228   .1640389    29.03   0.000     4.439602    5.082855
     firmage |   .0194495   .0038555     5.04   0.000     .0118902    .0270088
    firmage2 |   .0001008   .0000459     2.20   0.028     .0000109    .0001907
      gender |    -.07799   .0410329    -1.90   0.057    -.1584421     .002462
         age |  -.0136351   .0113263    -1.20   0.229    -.0358422     .008572
        age2 |   .0001217   .0001155     1.05   0.292    -.0001048    .0003482
      yrinco |   .0037037   .0010316     3.59   0.000      .001681    .0057264
             |
   education |
          0  |          0  (empty)
          2  |   .1131572   .0929838     1.22   0.224    -.0691536     .295468
          3  |    .106313   .0955127     1.11   0.266    -.0809563    .2935822
          4  |  -.0374574   .0967761    -0.39   0.699    -.2272037    .1522889
------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    50) =   957.43
  Prob > F      =   0.0000
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    50) =   957.43
  Prob > F      =   0.0000



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    50)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    50)
postxdeltatr |     957.43    0.0000 |     1096.21   0.0000 |      957.43

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=10.02    P-val=0.0015

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                    2168.06
Kleibergen-Paap Wald rk F statistic                               957.43

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,50)=        3.23     P-val=0.0783
Anderson-Rubin Wald test           Chi-sq(1)=      3.70     P-val=0.0545
Stock-Wright LM S statistic        Chi-sq(1)=      4.41     P-val=0.0358

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         51
Number of observations               N  =       3758
Number of regressors                 K  =         11
Number of endogenous regressors      K1 =          1
Number of instruments                L  =         11
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on mainethcoder

Number of clusters (mainethcoder) =     51            Number of obs =     3758
                                                      F( 11,    50) =     5.26
                                                      Prob > F      =   0.0000
Total (centered) SS     =  967.8062459                Centered R2   =   0.0057
Total (uncentered) SS   =  967.8062459                Uncentered R2 =   0.0057
Residual SS             =  962.2469273                Root MSE      =    .5362

---------------------------------------------------------------------------------
                |               Robust
   ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
----------------+----------------------------------------------------------------
postxdeltatrust |   .0824844   .0464542     1.78   0.082    -.0108216    .1757903
     postchange |  -.0647301   .0403166    -1.61   0.115    -.1457084    .0162483
        firmage |  -.0232746   .0056804    -4.10   0.000    -.0346841   -.0118651
       firmage2 |  -.0000106   .0001111    -0.10   0.925    -.0002336    .0002125
         gender |  -.1144188   .0900087    -1.27   0.210    -.2952067     .066369
            age |   .0049401   .0137552     0.36   0.721     -.022688    .0325681
           age2 |  -.0000835   .0001204    -0.69   0.491    -.0003254    .0001584
         yrinco |   .0008445   .0011285     0.75   0.458    -.0014222    .0031113
                |
      education |
             0  |          0  (empty)
             2  |   .2144293   .1041217     2.06   0.045     .0052948    .4235638
             3  |   .1742841   .1035664     1.68   0.099    -.0337352    .3823033
             4  |    .293476   .0925513     3.17   0.003     .1075812    .4793707
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             10.020
                                                   Chi-sq(1) P-val =    0.0015
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             2168.055
                         (Kleibergen-Paap rk Wald F statistic):        957.428
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         postxdeltatrust
Included instruments: postchange firmage firmage2 gender age age2 yrinco
                      2.education 3.education 4.education
Excluded instruments: postxtrustbf
Partialled-out:       _cons
                      nb: total SS, model F and R2s are after partialling-out;
                          any small-sample adjustments include partialled-out
                          variables in regressor count K
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       388           0         388     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd/died"

added macro:
               e(spec) : "IV"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3758        388
(e(noevents) = 388 added)

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
                                    \multicolumn{8}{c}{arsinh(Future patent applications)}                                                                                 
                                              (1)              (2)              (3)              (4)              (5)              (6)              (7)              (8)   
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
CEO's trust                                 0.241**          0.069            0.071            0.525*           0.512*           0.085*                                    
                                          (0.093)          (0.041)          (0.045)          (0.256)          (0.276)          (0.043)                                     
Post $\times$ Preceding CEO's trust                                                                                                              -0.078*                   
                                                                                                                                                (0.043)                    
Post $\times$ Change in trust                                                                                                                                      0.082*  
                                                                                                                                                                 (0.046)   
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sample                                    Retd 65       Retd 64-66       Retd 63-67    Died trans yr    Died trans yr+1        Retd/died        Retd/died        Retd/died   
Specification                                 OLS              OLS              OLS              OLS              OLS              OLS               RF               IV   
Event \& Year FEs                               X                X                X                X                X                X                X                X   
Baseline controls                               X                X                X                X                X                X                X                X   
Observations                                  914            2,287            3,442              253              353            3,758            3,758            3,758   
Events                                         93              232              348               34               46              388              388              388   
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
(11,137 missing values generated)
(207 missing values generated)
(2,304 observations deleted)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)
note: firmage is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)

HDFE Linear regression                            Number of obs   =        826
Absorbing 2 HDFE groups                           F(   9,     23) =       2.25
Statistics robust to heteroskedasticity           Prob > F        =     0.0568
                                                  R-squared       =     0.9328
                                                  Adj R-squared   =     0.9222
                                                  Within R-sq.    =     0.0304
Number of clusters (mainethcode) =         24     Root MSE        =     0.5558

                           (Std. err. adjusted for 24 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .2785619    .101939     2.73   0.012     .0676849    .4894388
     firmage |          0  (omitted)
    firmage2 |  -.0003139   .0002477    -1.27   0.218    -.0008262    .0001984
      gender |   -.387255   .3333285    -1.16   0.257    -1.076798    .3022876
         age |  -.0162345   .0464168    -0.35   0.730     -.112255     .079786
        age2 |   .0001233   .0004469     0.28   0.785    -.0008012    .0010479
      yrinco |   .0068471   .0039588     1.73   0.097    -.0013422    .0150364
             |
   education |
          2  |   .2765765   .1325422     2.09   0.048     .0023922    .5507608
          3  |   .4161425   .1674039     2.49   0.021      .069841    .7624439
          4  |   .4149474   .2104064     1.97   0.061    -.0203113    .8502062
             |
       _cons |   1.016275   1.457113     0.70   0.493    -1.997993    4.030543
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |        92           0          92     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd 65"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |        826         92
(e(noevents) = 92 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =      2,075
Absorbing 2 HDFE groups                           F(  10,     26) =      10.74
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.9373
                                                  Adj R-squared   =     0.9286
                                                  Within R-sq.    =     0.0102
Number of clusters (mainethcode) =         27     Root MSE        =     0.5348

                           (Std. err. adjusted for 27 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .0961995   .0575251     1.67   0.106    -.0220451     .214444
     firmage |  -.0185721   .0068635    -2.71   0.012    -.0326802    -.004464
    firmage2 |  -.0002208   .0001643    -1.34   0.191    -.0005585    .0001169
      gender |  -.2069808   .2329588    -0.89   0.382    -.6858344    .2718729
         age |  -.0040129   .0304425    -0.13   0.896    -.0665883    .0585626
        age2 |   .0000498    .000288     0.17   0.864    -.0005421    .0006418
      yrinco |   .0023988   .0027525     0.87   0.391    -.0032591    .0080567
             |
   education |
          2  |   .2212001   .1169899     1.89   0.070    -.0192761    .4616764
          3  |   .1570573   .1273067     1.23   0.228    -.1046254    .4187399
          4  |   .2282685   .1445182     1.58   0.126    -.0687929    .5253299
             |
       _cons |   1.847272   1.033094     1.79   0.085    -.2762831    3.970828
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       230           0         230     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd 64-66"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       2075        230
(e(noevents) = 230 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 6 iterations)

HDFE Linear regression                            Number of obs   =      3,130
Absorbing 2 HDFE groups                           F(  10,     27) =       3.85
Statistics robust to heteroskedasticity           Prob > F        =     0.0025
                                                  R-squared       =     0.9326
                                                  Adj R-squared   =     0.9237
                                                  Within R-sq.    =     0.0083
Number of clusters (mainethcode) =         28     Root MSE        =     0.5377

                           (Std. err. adjusted for 28 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .1103891   .0539688     2.05   0.051    -.0003457    .2211239
     firmage |  -.0195761   .0067406    -2.90   0.007    -.0334067   -.0057455
    firmage2 |  -.0000587   .0000769    -0.76   0.452    -.0002165    .0000991
      gender |  -.1413424   .1495974    -0.94   0.353    -.4482908    .1656061
         age |    .001472   .0253811     0.06   0.954    -.0506057    .0535496
        age2 |  -.0000295   .0002489    -0.12   0.907    -.0005401    .0004812
      yrinco |   .0027643   .0020051     1.38   0.179    -.0013498    .0068784
             |
   education |
          2  |   .2101798   .1162584     1.81   0.082    -.0283628    .4487224
          3  |   .1488129    .115726     1.29   0.209    -.0886373    .3862631
          4  |   .2890085   .1389171     2.08   0.047     .0039741    .5740428
             |
       _cons |   1.398008   .6300545     2.22   0.035     .1052431    2.690773
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       346           0         346     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd 63-67"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3130        346
(e(noevents) = 346 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 8 iterations)
note: firmage is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)

HDFE Linear regression                            Number of obs   =        222
Absorbing 2 HDFE groups                           F(   9,     10) =  174360.02
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.9364
                                                  Adj R-squared   =     0.9158
                                                  Within R-sq.    =     0.0886
Number of clusters (mainethcode) =         11     Root MSE        =     0.5692

                           (Std. err. adjusted for 11 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |    .918207   .4478201     2.05   0.067    -.0795983    1.916012
     firmage |          0  (omitted)
    firmage2 |   .0006071   .0005873     1.03   0.326    -.0007016    .0019157
      gender |  -.9039411   .9895091    -0.91   0.382    -3.108705    1.300822
         age |  -.0456464   .1226006    -0.37   0.717    -.3188176    .2275247
        age2 |    .000384    .000973     0.39   0.701     -.001784    .0025521
      yrinco |  -.0341225   .0215333    -1.58   0.144    -.0821016    .0138566
             |
   education |
          2  |   .5568091   .2615073     2.13   0.059    -.0258656    1.139484
          3  |   .3105895   .2168532     1.43   0.183    -.1725897    .7937686
          4  |   .8420884   .2563796     3.28   0.008     .2708391    1.413338
             |
       _cons |  -1.533607   6.852232    -0.22   0.827    -16.80133    13.73412
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |        34           0          34     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Died trans yr"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |        222         34
(e(noevents) = 34 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 8 iterations)
note: firmage is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)

HDFE Linear regression                            Number of obs   =        312
Absorbing 2 HDFE groups                           F(   9,     11) =    1310.84
Statistics robust to heteroskedasticity           Prob > F        =     0.0000
                                                  R-squared       =     0.9328
                                                  Adj R-squared   =     0.9146
                                                  Within R-sq.    =     0.0615
Number of clusters (mainethcode) =         12     Root MSE        =     0.5272

                           (Std. err. adjusted for 12 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .8949164   .3309323     2.70   0.021     .1665393    1.623293
     firmage |          0  (omitted)
    firmage2 |   .0005485     .00049     1.12   0.287      -.00053    .0016269
      gender |  -.6122023   .2210695    -2.77   0.018    -1.098773   -.1256317
         age |  -.0663652   .0945099    -0.70   0.497    -.2743801    .1416497
        age2 |   .0005369   .0007645     0.70   0.497    -.0011458    .0022197
      yrinco |   -.027954   .0150396    -1.86   0.090    -.0610559    .0051479
             |
   education |
          2  |   .4689882   .1056149     4.44   0.001     .2365315     .701445
          3  |   .1194852    .121645     0.98   0.347    -.1482536     .387224
          4  |   .5106982   .1857923     2.75   0.019     .1017722    .9196243
             |
       _cons |  -1.165765   4.821837    -0.24   0.813    -11.77856    9.447026
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |        46           0          46     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Died trans yr+1"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |        312         46
(e(noevents) = 46 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =      3,409
Absorbing 2 HDFE groups                           F(  10,     27) =       4.23
Statistics robust to heteroskedasticity           Prob > F        =     0.0013
                                                  R-squared       =     0.9319
                                                  Adj R-squared   =     0.9227
                                                  Within R-sq.    =     0.0087
Number of clusters (mainethcode) =         28     Root MSE        =     0.5394

                           (Std. err. adjusted for 28 clusters in mainethcode)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
    trust_sd |   .1303213   .0514737     2.53   0.017      .024706    .2359366
     firmage |  -.0216175   .0065879    -3.28   0.003    -.0351347   -.0081002
    firmage2 |  -.0000238   .0000724    -0.33   0.745    -.0001723    .0001247
      gender |  -.1430628   .1374527    -1.04   0.307    -.4250924    .1389668
         age |   .0006919   .0139701     0.05   0.961    -.0279724    .0293562
        age2 |  -.0000275   .0001303    -0.21   0.834    -.0002948    .0002398
      yrinco |   .0022718   .0018404     1.23   0.228    -.0015043     .006048
             |
   education |
          2  |   .2102925    .105014     2.00   0.055    -.0051785    .4257635
          3  |   .1571063   .1059399     1.48   0.150    -.0602643    .3744769
          4  |   .3148411    .126112     2.50   0.019     .0560806    .5736017
             |
       _cons |   1.327314   .4679772     2.84   0.009     .3671042    2.287524
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       386           0         386     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd/died"

added macro:
               e(spec) : "OLS"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3409        386
(e(noevents) = 386 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

HDFE Linear regression                            Number of obs   =      3,409
Absorbing 2 HDFE groups                           F(  11,     50) =       3.91
Statistics robust to heteroskedasticity           Prob > F        =     0.0004
                                                  R-squared       =     0.9318
                                                  Adj R-squared   =     0.9226
                                                  Within R-sq.    =     0.0071
Number of clusters (mainethcoder) =         51    Root MSE        =     0.5400

                          (Std. err. adjusted for 51 clusters in mainethcoder)
------------------------------------------------------------------------------
             |               Robust
ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
postxtrustbf |  -.1160841   .0507519    -2.29   0.026    -.2180222    -.014146
  postchange |   .5342212   .2489739     2.15   0.037     .0341424      1.0343
     firmage |   -.022367   .0060006    -3.73   0.000    -.0344196   -.0103144
    firmage2 |  -3.05e-06   .0001137    -0.03   0.979    -.0002314    .0002253
      gender |  -.1572696   .1178762    -1.33   0.188    -.3940309    .0794918
         age |   .0003082   .0184361     0.02   0.987    -.0367218    .0373382
        age2 |  -.0000441   .0001595    -0.28   0.784    -.0003645    .0002763
      yrinco |   .0023025    .001304     1.77   0.084    -.0003167    .0049216
             |
   education |
          2  |   .2255174    .100334     2.25   0.029     .0239907    .4270441
          3  |   .1747482   .1047733     1.67   0.102    -.0356952    .3851915
          4  |   .3111261   .0990862     3.14   0.003     .1121055    .5101466
             |
       _cons |   2.092607   .6367439     3.29   0.002     .8136691    3.371545
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       386           0         386     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd/died"

added macro:
               e(spec) : "RF"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3409        386
(e(noevents) = 386 added)
WARNING: Singleton observations not dropped; statistical significance is biased (link)
(MWFE estimator converged in 7 iterations)

First-stage regressions
-----------------------


First-stage regression of postxdeltatrust:

Statistics robust to heteroskedasticity and clustering on mainethcoder
Number of obs =                   3409
Number of clusters (mainethcoder) =     51
------------------------------------------------------------------------------
             |               Robust
postxdelta~t | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
postxtrustbf |  -.9285301   .0335571   -27.67   0.000    -.9943273    -.862733
  postchange |   4.690635   .1802553    26.02   0.000     4.337198    5.044071
     firmage |    .017766   .0041833     4.25   0.000     .0095635    .0259685
    firmage2 |   .0001128   .0000458     2.46   0.014     .0000229    .0002027
      gender |  -.0914275   .0441118    -2.07   0.038    -.1779199   -.0049351
         age |  -.0142692   .0130239    -1.10   0.273    -.0398059    .0112675
        age2 |   .0001245   .0001309     0.95   0.342    -.0001321    .0003811
      yrinco |   .0033648   .0011467     2.93   0.003     .0011165    .0056132
             |
   education |
          0  |          0  (empty)
          2  |   .1016713   .1015084     1.00   0.317    -.0973618    .3007044
          3  |   .0941877   .1051079     0.90   0.370    -.1119031    .3002784
          4  |  -.0606687   .1066473    -0.57   0.569    -.2697779    .1484404
------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    50) =   765.64
  Prob > F      =   0.0000
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    50) =   765.64
  Prob > F      =   0.0000



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    50)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    50)
postxdeltatr |     765.64    0.0000 |      887.16   0.0000 |      765.64

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=9.75     P-val=0.0018

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                    1878.03
Kleibergen-Paap Wald rk F statistic                               765.64

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,50)=        5.23     P-val=0.0264
Anderson-Rubin Wald test           Chi-sq(1)=      6.06     P-val=0.0138
Stock-Wright LM S statistic        Chi-sq(1)=      4.51     P-val=0.0338

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         51
Number of observations               N  =       3409
Number of regressors                 K  =         11
Number of endogenous regressors      K1 =          1
Number of instruments                L  =         11
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on mainethcoder

Number of clusters (mainethcoder) =     51            Number of obs =     3409
                                                      F( 11,    50) =     3.61
                                                      Prob > F      =   0.0009
Total (centered) SS     =  880.9204786                Centered R2   =   0.0083
Total (uncentered) SS   =  880.9204786                Uncentered R2 =   0.0083
Residual SS             =  873.6259294                Root MSE      =    .5396

---------------------------------------------------------------------------------
                |               Robust
   ash_f1allpat | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
----------------+----------------------------------------------------------------
postxdeltatrust |   .1250192   .0552152     2.26   0.028     .0141162    .2359222
     postchange |  -.0521983   .0537691    -0.97   0.336    -.1601967    .0558001
        firmage |  -.0245881   .0064363    -3.82   0.000    -.0375158   -.0116604
       firmage2 |  -.0000171   .0001139    -0.15   0.881     -.000246    .0002117
         gender |  -.1458394   .1197941    -1.22   0.229    -.3864529    .0947742
            age |   .0020921   .0179038     0.12   0.907    -.0338687     .038053
           age2 |  -.0000596    .000155    -0.38   0.702    -.0003708    .0002516
         yrinco |   .0018818   .0013199     1.43   0.160    -.0007693    .0045329
                |
      education |
             0  |          0  (empty)
             2  |   .2128066   .1028416     2.07   0.044     .0062431      .41937
             3  |   .1629729   .1058535     1.54   0.130    -.0496401    .3755859
             4  |   .3187108   .1036423     3.08   0.003     .1105391    .5268826
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              9.746
                                                   Chi-sq(1) P-val =    0.0018
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):             1878.028
                         (Kleibergen-Paap rk Wald F statistic):        765.640
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         postxdeltatrust
Included instruments: postchange firmage firmage2 gender age age2 yrinco
                      2.education 3.education 4.education
Excluded instruments: postxtrustbf
Partialled-out:       _cons
                      nb: total SS, model F and R2s are after partialling-out;
                          any small-sample adjustments include partialled-out
                          variables in regressor count K
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
     eventid |       386           0         386     |
        year |        13           1          12     |
-----------------------------------------------------+

added macro:
             e(sample) : "Retd/died"

added macro:
               e(spec) : "IV"

added macro:
                 e(FE) : "X"

added macro:
           e(controls) : "X"

         |        Observations
         |      total   distinct
---------+----------------------
 eventid |       3409        386
(e(noevents) = 386 added)

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
                                    \multicolumn{8}{c}{arsinh(Future patent applications)}                                                                                 
                                              (1)              (2)              (3)              (4)              (5)              (6)              (7)              (8)   
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
CEO's trust                                 0.279**          0.096            0.110*           0.918*           0.895**          0.130**                                   
                                          (0.102)          (0.058)          (0.054)          (0.448)          (0.331)          (0.051)                                     
Post $\times$ Preceding CEO's trust                                                                                                              -0.116**                  
                                                                                                                                                (0.051)                    
Post $\times$ Change in trust                                                                                                                                      0.125** 
                                                                                                                                                                 (0.055)   
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sample                                    Retd 65       Retd 64-66       Retd 63-67    Died trans yr    Died trans yr+1        Retd/died        Retd/died        Retd/died   
Specification                                 OLS              OLS              OLS              OLS              OLS              OLS               RF               IV   
Event \& Year FEs                               X                X                X                X                X                X                X                X   
Baseline controls                               X                X                X                X                X                X                X                X   
Observations                                  826            2,075            3,130              222              312            3,409            3,409            3,409   
Events                                         92              230              346               34               46              386              386              386   
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------

. 
. cap log close
